Arthritis & Rheumatism, Volume 62,
November 2010 Abstract Supplement

Abstracts of the American College of
Rheumatology/Association of Rheumatology Health Professionals
Annual Scientific Meeting
Atlanta, Georgia November 6-11, 2010.

Risk Factors in Different Subsets of Rheumatoid Arthritis.

Lundberg3,  Karin, Bengtsson1,  Camilla, Israelsson4,  Lena, Pollak-Dorocic4,  Iskra, Padyukov4,  Leonid, Alfredsson1,  Lars, Malmstrom4,  Vivianne

Institute for Environmental Medicine, Karolinska Institutet
Karolinska University Hospital, Stockholm, Sweden
Reumatology Unit, Karolinska Institutet, Stockholm, Sweden
Reumatology Unit, Karolinska Institutet


Risk factors for rheumatoid arthritis (RA), such as HLA-DRB1'shared epitope' (SE) alleles, PTPN22 and cigarette smoking, associate with anti-CCP positive disease. By subdividing CCP positive patients based on reactivity to CEP-1, the immunodominant B cell epitope on citrullinated alpha-enolase, we have previously shown that SE, PTPN22 and smoking mainly constitute risk factors for CEP-1 positive, rather than CCP positive RA. Here, we investigate 5 RA subsets, grouped based on reactivity to CCP, CEP-1 and a citrullinated vimentin peptide (CitVim), representing another anti-citrullinated protein/peptide antibody (ACPA)-specificity.


Antibodies were analysed in serum from 1027 RA patients from the Epidemiological Investigation of RA (EIRA), a Swedish population-based case-control study. An ELISA was set up for detection of anti-CitVim antibodies, using a biotinylated CitVim peptide (amino acid 60–75). One hundred and fifty healthy controls were used to determine the 95th percentile cut-off. A positive and a negative control and a standard of pooled anti-CitVim positive serum were included on each plate. Anti-CCP was measured using the Immunoscan CCPlus kit (Euro-Diagnostica) and anti-CEP-1, using an in-house ELISA previously described (Mahdi et al, Nature Genetics, 2009). Different RA subsets were compared with regard to risk factors by calculating odds ratios (OR) with 95% confidence interval by means of logistic regression.


Patients were divided into 5 subsets: CEP-1+/CitVim+ (41%); CEP-1+/CitVim- (15%); CEP-1-/CitVim+ (11%) (the majority of patients in these subsets were anti-CCP positive and not divided further based on CCP status); CEP-1-/CitVim-/CCP+ (18%) and CEP-1-/CitVim-/CCP- (15%). SE, PTPN22 and smoking each showed strongest association in the CEP-1+/CitVim+ subset, intermediate association in the single positive subsets and low association in the ACPA negative subset. The combined effect of risk factors showed a similar pattern, with an OR of 52 in the double positive subset, 16 in the CEP-1 single positive, 5 in the CitVim single positive, 6 in the CCP single positive, and 2 in the triple negative. Additionally, antibody levels were higher in CEP-1+/CitVim+ patients compared to single positive patients.


Our data show that the anti-CEP-1 and the anti-CitVim antibody response only partly overlap, and that SE, PTPN22 and smoking predispose to the development of ACPA with multiple reactivities, rather than one single specificity. Interestingly, the CEP-1-/CitVim-/CCP+ subset still showed a strong association with SE (OR: 12), suggesting that other ACPA fine-specificities, for example citrullinated fibrinogen or collagen type II, could be found in this subset.

To cite this abstract, please use the following information:
Lundberg, Karin, Bengtsson, Camilla, Israelsson, Lena, Pollak-Dorocic, Iskra, Padyukov, Leonid, Alfredsson, Lars, et al; Risk Factors in Different Subsets of Rheumatoid Arthritis. [abstract]. Arthritis Rheum 2010;62 Suppl 10 :2118
DOI: 10.1002/art.29882

Abstract Supplement

Meeting Menu